Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Red-Teaming (ASR) on DANGEROUSQA

0ASR

GPT-4

-3.6621.04545.7570.455Aug 18, 2023Jan 20, 2024Jun 23, 2024Nov 25, 2024Apr 29, 2025Oct 1, 2025Mar 6, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2023.08
0----
2023.08
0----
2023.08
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2026.03
0----
2023.08
0.5----
2023.08
1.5----
2023.08
2.5----
2023.08
2.6----
2023.08
2.7----
2026.03
3.5----
2023.08
5----
2026.03
5----
2026.03
6.5----
2023.08
9.5----
2023.08
10.2----
2026.03
13----
2026.03
14.25----
2026.03
16.5----
2026.03
17----
2026.03
19----
2026.03
28----
2026.03
33.5----
2023.08
36.7----
2023.08
46.5----
2023.08
48.5----
2023.08
49----
2023.08
53.2----
2023.08
57----
2023.08
63----
2023.08
72.2----
2023.08
73.6----
2023.08
75.5----
2023.08
81.5----
2023.08
82.5----
2023.08
83.5----
2023.08
86----
2023.08
86.5----
2023.08
87----
2023.08
90----
2023.08
90.5----
2023.08
91.5----
2023.08
-12.2---
2023.08
-24.7---
2023.08
-46.2---
2023.08
-49---
2023.08
-49---
2023.08
-58.7---
2023.08
-46.5---
2023.08
-82.7---
2023.08
-44.1---
2023.08
-49.5---
2023.08
-21.7--65.1
2023.08
-24.400.00572.8
2023.08
-450.0270.4983.5
2023.08
-47.70.0250.53287.5
2023.08
-52.30.0260.6391.5
2023.08
-590.1020.75591.5
2023.08
-47.30.0950.46586
2023.08
-82.60.7220.8689.6
2023.08
-42.10.0150.48576.5
2023.08
-49.20.050.5785.5